Finding point correspondences in motion sequences preserving affine structure
In this paper the problem of computing the point correspondences in a sequence of time-varying images of a 3D object undergoing nonrigid (affine) motion is addressed. It is assumed that the images are obtained through affine projections. The correspondences are established only from the analysis of...
| Huvudupphovsmän: | , , |
|---|---|
| Materialtyp: | Journal article |
| Språk: | English |
| Publicerad: |
Elsevier
1997
|
| _version_ | 1826317564954804224 |
|---|---|
| author | Sudhir, G Banerjee, S Zisserman, A |
| author_facet | Sudhir, G Banerjee, S Zisserman, A |
| author_sort | Sudhir, G |
| collection | OXFORD |
| description | In this paper the problem of computing the point correspondences in a sequence of time-varying images of a 3D object undergoing nonrigid (affine) motion is addressed. It is assumed that the images are obtained through affine projections. The correspondences are established only from the analysis of the unknown 3D affine structure of the object, without making use of any attributes of the feature points. It is shown that it is possible to establish the point correspondences uniquely (up to symmetry) in the sense that they yield a unique affine structure of the object and that the computation is possible in polynomial time. Two different algorithms for computing the point correspondences are presented. Results on various real image sequences, including a sequence containing independently moving objects, demonstrate the applicability of the structure based approach to motion correspondence.
|
| first_indexed | 2025-03-11T16:55:55Z |
| format | Journal article |
| id | oxford-uuid:e5f07d20-e2f5-4fd2-9fdd-a2d3c225c999 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2025-03-11T16:55:55Z |
| publishDate | 1997 |
| publisher | Elsevier |
| record_format | dspace |
| spelling | oxford-uuid:e5f07d20-e2f5-4fd2-9fdd-a2d3c225c9992025-02-20T14:15:31ZFinding point correspondences in motion sequences preserving affine structureJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:e5f07d20-e2f5-4fd2-9fdd-a2d3c225c999EnglishSymplectic ElementsElsevier1997Sudhir, GBanerjee, SZisserman, AIn this paper the problem of computing the point correspondences in a sequence of time-varying images of a 3D object undergoing nonrigid (affine) motion is addressed. It is assumed that the images are obtained through affine projections. The correspondences are established only from the analysis of the unknown 3D affine structure of the object, without making use of any attributes of the feature points. It is shown that it is possible to establish the point correspondences uniquely (up to symmetry) in the sense that they yield a unique affine structure of the object and that the computation is possible in polynomial time. Two different algorithms for computing the point correspondences are presented. Results on various real image sequences, including a sequence containing independently moving objects, demonstrate the applicability of the structure based approach to motion correspondence. |
| spellingShingle | Sudhir, G Banerjee, S Zisserman, A Finding point correspondences in motion sequences preserving affine structure |
| title | Finding point correspondences in motion sequences preserving affine structure |
| title_full | Finding point correspondences in motion sequences preserving affine structure |
| title_fullStr | Finding point correspondences in motion sequences preserving affine structure |
| title_full_unstemmed | Finding point correspondences in motion sequences preserving affine structure |
| title_short | Finding point correspondences in motion sequences preserving affine structure |
| title_sort | finding point correspondences in motion sequences preserving affine structure |
| work_keys_str_mv | AT sudhirg findingpointcorrespondencesinmotionsequencespreservingaffinestructure AT banerjees findingpointcorrespondencesinmotionsequencespreservingaffinestructure AT zissermana findingpointcorrespondencesinmotionsequencespreservingaffinestructure |